Need an audio clean done

[Greg Miller]

Originally Posted by Jim Andrada

I'd have to say that the original clip has more harmonics with significant energy marching into the frequency stratosphere than I can remember seeing in a long time
{snip}
Was it recorded at the base of a broadcast tower or next to a radar installation - or were the mics hidden in a microwave oven:-)

Jim, the buzz consists of numerous exact harmonics of 300 Hz, first noticeable at 1,200 Hz and running up through 12,000 Hz. So as I said earlier, the exact 300 Hz spacing makes it easy to remove the harmonics with a bunch of notch filters; and that does not create artifacts, the way a "noise sample" approach does.

There must have been some sort of very exact 300 Hz signal present in the equipment or nearby. I do not think it's related to a dimmer or other power line system, because there would be other harmonics of 50 Hz in there. Offhand I can't think of any equipment I've worked with that is running a precise 300 Hz clock signal... can you?

The repetitive impulse noise might be crosstalk from a display somewhere. Some of the early MP3 recorders had this problem, because of insufficient shielding, or ground loops, between the input stages and the display circuitry. The cheap "Wristwatch Spy Recorders" also produce noise like this... maybe this track was recorded on one of those (still not in the right location).

Recently, I heard this sort of impulse noise when I was testing a mic made by "Sound Professionals." I discovered that if the mic wire was anywhere near the display on my DR-07 or DR-03, it would pick up noise like that. The source of the noise is apparently the strobe signal from the display. The reason that mic picked it up is because "Sound Professionals" uses unshielded headphone wires on many of their microphones! Appalling! When I tried the same test with a "real" microphone, using shielded cable, of course the noise was gone.

[Jim Andrada]

Hi Greg

Yes - that's the thing that's bothering me - I can't think of anything that would be at exactly 300Hz.

I would also tend to think most/all of the "trash" would not have been present if proper cabling had been used.

However, even absent the noise, I think this would be hard to salvage. Wrong/improperly placed mics for sure. Poor recording space.

[Greg Miller]

Originally Posted by Jim Andrada

I would also tend to think most/all of the "trash" would not have been present if proper cabling had been used.

Hi Jim,

Good point, and I really didn't think about that. The highest audio was only about -10dB below FS, so the problem isn't caused by low mic gain as the OP suggested. In fact if recording gain had been 10dB higher, the noise would probably have been 10dB higher, as well.

In fact, maybe the recording gain was too high, and that's what caused the noise level to be so high. Clearly nobody in this clip is anywhere near being "on mic" so maybe that caused the camera op to set the gain too high, increasing the noise level, too.

I sure would like to know the source of the 300Hz pulse. IIRC the old rotary VCRs had a sync pulse that was 4X the nominal frame rate; but I can't imagine what would be using 300Hz these days. And would the freq. be higher or lower on the "defrost" setting?

[Jim Andrada]

Yeah, this one has me stumped - if it were a motor arcing it would have to be running at 18k RPM - I even Googled around looking for 300Hz.

Whatever it is, it's well regulated and consistent - wonder if it could be the head motor if the camera was using tape - not sure how fast that spins - I think 18k might be too high.

The other surprising thing (to me at least) was how high in the harmonic range there was significant energy. I think most hum has pretty low energy in the upper harmonics.

'Tis a mystery - MUST have been the defrost cycle.

I'm also thinking that the gain must have been cranked up about as far as it would go.

The other thing that has me scratching what's left of my hair is what kind of group this was. I heard a few different accents on the English speech, something completely unintelligible, and I clearly heard somebody speaking Japanese.

[Greg Miller]

I very much doubt a motor arcing. First of all, the arcing would not be constant, it would be continuously changing. So even if the fundamental frequency was steady, the level of the various harmonics would be up and down randomly. But as I think about it, I doubt that the fundamental would be that steady, because of minor dimensional variations in the size of the commutator segments. Nah, not an arcing motor.

It's not hum based. "Pure" hum would be just the fundamental. You get harmonics on power line "hum" because of changing imbalance on the 3-phase distribution system, as well as higher harmonics from switched loads like dimmers, variable speed motors, etc. These tend to be odd-order harmonics, so they'd be 300, 900, 1500, etc. On the other hand, if I recall correctly, a triangle wave or a narrow pulse contains "all" harmonics... "all" meaning within the bandwidth of the system in question. So apparently something in the recording chain rolls off up around 12kHz. (Actually, I'm a little unclear about why the first 3 harmonics -- 300, 600, 900 -- seem to be missing. But depending on how the noise was coupled into the recording chain, I guess the coupling at lower frequencies might be much more lossy than coupling at higher freqs.)

Hah! I also Googled things like "300 Hz sync" and "300 Hz clock" but found nothing that seemed even vaguely relevant. Interestingly, I googled "300 Hz sync pulse" about 2 minutes after my initial reply here, and this thread came up at the first hit! How do they do that?????

It's much too fast for vertical sweep, and too slow for horizontal sweep, on a typical display. Surely nobody would use a 300Hz inverter to step up battery voltage to 48v phantom voltage... they'd want to use something far above the audio range. So I keep coming back to something like the pulse from the sync generator on a motor, or a reference pulse for a motor somewhere.

I hope the OP will fess up and tell us what sort of equipment was used to create this recording, so we can do a little more research. In fact complete silence from the OP makes me wonder whether he intentionally created this mix just to pull our legs. (OK, so I always was a sucker for conspiracy theories.)

Otherwise, my hunch is that this was recorded in a 3-room apartment, with an English soap playing on a TV in one room, a Japanese soap playing on a TV in the second room, and a "Spy Watch" recorder hidden under a pillow in the third room.

[Jim Andrada]

Damn - makes me feel like Sherlock Ohms on a bad hair day.

Just can't come up with anything remotely plausible to explain this mixed salad of noise.

Maybe it's an international student lounge set up inside a giant industrial microwave oven. And they're all being defrosted.

By the way, came up with an interesting site whilst googling around. (I remember using 2 inch tapes a million years ago, although not for video - they were for radar signature analysis and the reel was about a meter in diameter and mounted on the rather enormous reel to reel drive with a little forklift.)

Scan